DE69018920T2 - USE OF DNA FRAGMENTS ENCODING A SIGNAL PEPTIDE FOR SECRETION OF MATURE PROTEINS IN RECOMBINANT YEARS, EXPRESSION CASSETTES, TRANSFORMED YEARS AND METHOD FOR PRODUCING THESE PROTEINS. - Google Patents

Abstract

The invention relates to new DNA fragments and to their application as a DNA coding fragment for a signal peptide which can be used for the secretion of proteins, said peptide including a sequence of amino-acids which show a degree of correspondence of at least 60% with the sequence of amino-acids (I) or (II), preferably with the sequence (II). The sequences (I) and (II) are as follows: (I) Arg-Phe-Ser-Thr-Thr-Leu-Ala-Thr-Ala-Ala-Thr-Ala-Leu-Phe-Phe-Thr-Ala-Ser-Gln. (II) Arg-Phe-Ser-Thr-Thr-Leu-Ala-Thr-Ala-Ala-Thr-Ala-Leu-Phe-Phe-Thr-Ala-Ser-Gln-Val-Ser-Ala.

Description

The present invention relates to novel DNA fragments and their use as DNA fragments encoding a signal peptide suitable for the secretion of heterologous proteins by (animal or plant) eukaryotic cells or prokaryotic cells (bacteria), in particular yeasts, e.g. strains of Saccharomyces.

When producing a heterologous protein using recombinant DNA techniques, one of the most frequently pursued objectives is the recovery (production) of a product which is secreted into the culture medium of the cells which synthesize the heterologous protein. In particular, in the case of producing a protein of industrial interest which is to be produced in large quantities, it is desirable that the protein be produced in mature form, i.e. free of any amino acid or additional peptide sequence which has remained fused to the protein, in order to retain the desired properties. In addition, it may be advantageous for it to be secreted into the culture medium in order to facilitate the recovery and purification operations.

The proteins secreted by a cell, especially a eukaryotic cell, are generally synthesized in the form of a polypeptide precursor, which includes a fragment corresponding to the mature protein (in the active form) and an N-terminal fragment called the "pre"fragment. Fragment, also called signal peptide, which interferes with the mechanism of secretion of the protein by the cell. In addition, this polypeptide precursor may contain one or more additional fragments, called "pro" fragments. In the latter case, the polypeptide precursor is called a "pre-pro" precursor or first precursor. A "pro" fragment is inserted in the majority of cases between the signal peptide and the fragment corresponding to the mature protein, although this is not an absolute rule.

A signal peptide initiates (i) the insertion of the protein into the cell membrane, (ii) the translocation of the protein across the cell membrane, or (iii) the entry of the protein into the endoplasmic reticulum of the cell for secretion of the protein through the endoplasmic reticulum. Once the signal peptide has performed its task, it is usually removed by proteolytic cleavage to release a mature protein or a second precursor, called a "pro" precursor, which, like the first precursor, has no biological activity or does not have the full activity of the mature protein.

A "pro" fragment is useful in that it blocks or modifies the activity of the protein, thus allowing the cell to be protected against the possible toxic effects of the protein or the protein to be protected against possible modifications or degradation reactions. It can also interfere to some extent with the secretion mechanism. At the end of the secretion process, the "pro" fragment of the second precursor is removed by proteolytic cleavage to release a mature protein (in the active form).

At the junction between the "pre" fragment and the "pro" fragment as well as at the junction between the "pro" fragment and the fragment corresponding to the mature protein, there must be a proteolytic cleavage site (cutting site) which is defined by one of the Proteases of the cell in which the protein is synthesized. This proteolytic cleavage site generally consists of a sequence of two or three or more amino acids (hereinafter referred to as the "proteolysis sequence") which is accessible to the protease on the "pro" precursor and which is not accessible when present on the fragment corresponding to the mature protein. A priori, three cases are possible, which are explained below with reference to the connection of the "pre" and "pro" fragments:

- the protease cuts at the head of the sequence and consequently the proteolysis sequence must be read on the "pro" fragment;

- the protease cuts at the tail of the sequence and consequently the proteolysis sequence must be read on the "pre" fragment;

- the protease cuts in the middle of the sequence and consequently the proteolysis sequence must be read across the "pre" and "pro" fragments.

These considerations also apply in a similar way to the cleavage site (cutting site) located at the junction between the "pro" fragment and the fragment corresponding to the mature protein.

For the construction of synthetic precursors using genetic engineering techniques, it may be necessary to use natural fragments (i.e. those found in nature) and to insert a new proteolytic cleavage site (cleavage site) or certain amino acids to create a new proteolytic cleavage site (cleavage site) in association with the fragments at a good (suitable) location, whereby this new cleavage site (cleavage site) is recognized with certainty by one of the proteases of the cell in which the synthetic precursor is expressed.

An example of this type of synthesis and secretion described above is illustrated using the α-sexual pheromone of the yeast S. cerevisiae, also called factor a (which is encoded by the gene MFal or MFa2). Factor α is in fact synthesized in the form of the "pre-pro" precursor, as described by Kujan and Herskowitz in "Cell" (1982) 30:933. The amino acid sequence of the "pre" fragment of the precursor of factor α is Met Arg Phe Pro 5er Ile Phe Thr Ala Val Leu Phe Ala Ala 5er 5er Ala Leu Ala, while that of the "pro" fragment of the precursor of factor α is is Ala Pro Val Asn Thr Thr Thr Glu Asp Glu Thr Ala Gln Ile Pro Ala Glu Ala Val Ile Gly Tyr Ser Asp Leu Glu Gly Asp Phe Asp Val Ala Val Leu Pro Phe Ser Asn Ser Thr Asn Asn Gly Leu Leu Phe Ile Asn Thr Thr Ile Ala 5er Ile Ala Ala Lys Glu Glu Gly Val Ser Leu Asp.

Surprisingly, it has now been found that the N-terminal end of the precursor of a yeast enzyme, whose N-terminal end has the following amino acid sequence: Arg-Phe-Ser-Thr-Thr-Leu-Ala-Thr-Ala-Ala-Thr-Ala-Leu-Phe-Phe-Thr-Ala-Ser-Gln-Val-Ser-Ala, can be used as a signal peptide for the secretion of heterologous proteins, as can various variants of this N-terminal end. A "heterologous protein" is a protein that is not naturally produced by the host cell or that is encoded by a sequence that does not originate from the host cell.

Accordingly, the present invention relates to an isolated DNA fragment encoding a peptide whose amino acid sequence has a degree of homology of at least 60%, preferably at least 80%, with the amino acid sequences (I) or (II) given below, preferably with the sequence (II):

(I) Arg-Phe-Ser-Thr-Thr-Leu-Ala-Thr-Ala-Ala-Thr-Ala-Leu-Phe-Phe-Thr- Ala-Ser-Gln.

(II) Arg-Phe-Ser-Thr-Thr-Leu-Ala-Thr-Ala-Ala-Thr-Ala-Leu-Phe-Phe-Thr- Ala-Ser-Gln-Val-Ser-Ala.

An "isolated DNA fragment" is understood to mean a DNA fragment whose 3' end is not covalently linked to a DNA fragment encoding an enzyme having ß-1,3-glucanase activity, as described in particular by Klebl & Tanner in "J. Bact", Nov. 1989, 171: 6259.

A DNA fragment according to the invention encodes in particular a peptide having the following amino acid sequence (III):

(III) R₁-R₂-R₃-Thr-Thr-R₄-Ala-Thr-Ala-Ala-Thr-Ala-Leu-Phe- 1 5 10 Phe-Thr-Ala-R5-R₆ 15 15

where:

R1 is an amino acid selected from Arg and Lys,

R₂ and R₆ are each an amino acid, independently selected from

Ala, Asn, Cys, Gln, Gly, His, Ile, Leu, Met, Phe, Pro, Ser, Thr, Trp, Tyr and Val

R₃ and R₅ each represent an alkanoic acid independently selected from Asp, Gly, Asn, Pro and 5er and

R4 is an amino acid selected from Val, Leu, Ala, Cys, Phe, Ile and Met.

A DNA fragment according to the invention preferably encodes a peptide having the following amino acid sequence (IV):

(IV) R₁-R₂-R₃-Thr-Thr-R₄-Ala-Thr-Ala-Ala-Thr-Ala-Leu-Phe-Phe-Thr- Ala-R₅-R₆-R₆

where:

R1 is an amino acid selected from Arg and Lys,

R₂ and R₆ are each an amino acid, independently selected from

Ala, Asn, Cys, Gln, Gly, His, Ile, Leu, Met, Phe, Pro, Ser, Thr, Trp, Tyr and Val

R₃ and R₅ each represent an amino acid, independently selected from

Asp, Gly, Asn, Pro and Ser,

R4 is an amino acid selected from

Val, Leu Ala, Cys, Phe, Ile and Met,

R7 is a proteolysis sequence.

R�7 is preferably a proteolysis sequence R�8-R�9--

R₁₀, in which:

R�8 is an amino acid selected from

Ala, Val, Ser, Cys, Gly, Ile, Leu and Thr,

R�9 is an amino acid selected from

Ala, Arg, Cys, Gln, Gly, His, Ile, Leu, Met, Phe, Pro, Ser, Thr, Trp, Tyr and Val and

R₁₀ is an amino acid selected from

Ala, Cys, Gly, Leu, Pro, Gln, Ser and Thr.

According to the invention, a DNA fragment preferably encodes a peptide having an amino acid sequence that is selected from the following amino acid sequences (V) (VI), (VII) and (VIII):

(V) Arg-Phe-Ser-Thr-Thr-Leu-Ala-Thr-Ala-Ala-Thr-Ala-Leu-Phe-Phe-Thr- Ala-Ser-Gln

(VI) Arg-Phe-Ser-Thr-Thr-Val-Ala-Thr-Ala-Ala-Thr-Ala-Leu-Phe-Phe-Thr- Ala-Scr-Gln

(VII) Arg-Phe-Ser-Thr-Thr-Leu-Ala-Ala-Thr-Ala-Leu-Phe-Phe- Thr-Ala-Ser-Gln-R�7;

wherein R�7 is as defined above,

(VIII) Arg-Phe-Ser-Thr-Thr-Val-Ala-Thr-Ala-Ala-Thr-Ala-Leu-Phe-Phe- Thr-Ala-Ser-Gln-R�sub7;

wherein R7 is as defined above.

R₇ particularly preferably represents a sequence in which R₈ is Val, R₇ is Ser and R₁₀ is Ala.

A DNA fragment according to the invention most preferably has one of the following sequences (IX), (X), (XI) and (XII) as nucleotide sequence:

The peptides encoded by the DNA fragments of the invention contain a hydrophobic region having an α-helix structure between the amino acid in position 3 and the amino acid in position 18. This structure helps to make them suitable for use as a signal peptide. It is known that the hydrophobic portion of the signal sequences consists predominantly of the amino acids Ala, Cys, Phe, Ile, Leu, Met, Val. It is also possible to provide certain modifications of amino acids which do not cause any alteration in the ability of the signal peptide to play its role.

In another aspect, the present invention therefore relates to the use of a DNA fragment according to the invention as a DNA fragment encoding a signal peptide suitable for the secretion of a heterologous protein by a host cell in which the heterologous protein is synthesized. In general, the invention can be carried out in a prokaryotic cell or in a eukaryotic cell, preferably in the latter. The eukaryotic cell can be, for example, a mammalian cell or a yeast cell. The secretion of a heterologous protein with the aid of a signal peptide encoded by a fragment according to the invention is preferably carried out by a yeast cell, for example of the genus Saccharomyces, in particular the species S. cerevisiae.

If the DNA fragments code for a signal peptide, the DNA fragments according to the invention are of course preceded by a translation initiation codon, generally a codon that codes for a methionine, in particular an ATG.

The DNA fragments according to the invention can be used for the construction of a cassette for the expression of a protein, which is preceded by the initiation codon alone or in combination with other components, for example a "pro" fragment. These DNA fragments can be produced by chemical synthesis using an oligonucleotide synthesizer according to a method generally known to those skilled in the art.

The invention also relates to a cassette for the expression of a heterologous protein, which comprises a DNA fragment according to the invention as a DNA fragment encoding the signal peptide of said heterologous protein. In particular, the expression cassette according to the invention, which contains the information required for the secretion of a mature heterologous protein, comprises in succession at least:

a) a DNA fragment containing transcription and translation initiation signals,

b) a DNA fragment according to the invention,

c) a DNA fragment encoding a mature heterologous protein (with a translation termination codon).

In a first variant, fragment (b) can be directly fused in phase with fragment (c), provided that at the junction between fragment (b) and fragment (c) there is a DNA sequence encoding a proteolytic cleavage site (cutting site) to allow the release of a mature protein at the end of the expression and secretion process.

Preferably, the DNA sequence encoding a proteolytic cleavage site (cleavage site) is read onto fragment (b).

In a second variant, an expression cassette according to the invention also comprises a DNA fragment (b') that codes for a peptide fragment "pro". This fragment (b') is fused in phase with the fragment (b) and the fragment (c), at the connection point between the fragments (b) and (b') on the one hand and the fragments (b') and (c) on the other hand there is a DNA sequence that codes for a proteolytic cleavage site (cleavage site).

Numerous fragments (b') can be used in the cassettes according to the invention. In particular, various fragments (b') can be constructed synthetically. As an example, the construction of a synthetic fragment (b') starting from the DNA sequence coding for the "pro" fragment of the precursor of factor α is described below. This sequence can be used in whole or in part. In a specific embodiment, the sequence that has been deleted from the section coding for the amino acid in position 3 to 42 on the fragment "pro", i.e. the sequence coding for Ala Pro Gly Leu Leu Phe Ile Asn Thr Thr Ile Ala Ser Ile Ala Ala Lys Glu Glu Gly Val Ser Leu Asp, is used. To form the fragment (b'), this last-mentioned DNA sequence is followed by a sequence coding for (i) a peptide comprising a proteolytic cleavage site (cleavage site), or (ii) a proteolytic cleavage site (cleavage site), the last-mentioned embodiment being preferred. The last-mentioned cleavage site (cutting site) is in particular Lys-Arg or Arg-Arg, which is recognized by the yeast endopeptidase yscF, which is encoded by the KEX2 gene and which cuts in the region of the C-terminal end of the dipeptide Lys-Arg or Arg-Arg. In summary, a fragment (b') encodes in particular Ala Pro Gly Leu Leu Phe Ile Asn Thr Thr Ile Ala Ser Ile Ala Ala Lys Glu Glu Gly Val Ser Leu Asp Lys Arg.

The sequence (a) comprises in particular a functional promoter in the cell in which it is desired to synthesize the heterologous protein encoded by the fragment (c), preferably a functional promoter in yeast. For example, yeast-based promoters can be used whose functionality has been confirmed by the transcription of genes encoding heterologous proteins, for example the PGK, ENO1, MFα1 promoters or inducible promoters such as PH05, GAL1. If elements of the yeast MFα1 gene are used in the expression cassette, the promoter of the MFα1 gene can be used, for example.

Finally, the expression cassettes can also comprise a DNA fragment (d) which comprises signals for the termination of transcription, preferably functional signals in yeast, for example that of the PGK gene.

In general, an expression cassette according to the invention can be introduced into a prokaryotic cell or into a eukaryotic cell, preferably into a eukaryotic cell, e.g. into a mammalian cell or yeast cell, with a yeast cell being particularly preferred. This introduction can be carried out by arranging the cassette in an autonomous replication plasmid or in a construction intended for integration for direct introduction into the genome of the yeast, so that in both cases a transformed cell is obtained.

If the plasmid is autonomous, it contains elements that ensure its division and replication, for example an origin of replication such as that of yeast plasmid 2u. In addition, the plasmid may contain selection elements, such as the URA3 or LEU2 gene, which ensure the complementation of yeast ura3 or leu2. In particular, it is advantageous to use the URA3 gene deleted from its promoter (URA3-d).

These plasmids may also include elements that ensure their replication in bacteria when the Plasmid must be a shuttle plasmid, for example an origin of replication such as that of pBR322, a selection marker gene such as Amp and/or other elements known to the person skilled in the art.

In accordance with the above, the present invention also relates to a cell which has been transformed by (i) a DNA fragment according to the invention or (ii) an expression cassette according to the invention which has either been inserted into a plasmid or integrated into the genome of the cell.

When the promoter is that of the MFα1 gene, the transformed yeast cell is preferably of the sexual type MATα. For example, a strain of the genotype ura3 or leu2 or another is used, complemented by the plasmid in order to ensure the maintenance of the plasmid in the yeast by an appropriate selection pressure.

Finally, the present invention relates to a process for producing a heterologous protein, which is characterized in that a cell according to the invention is cultivated and in that said protein is recovered (separated) from the culture medium. This process is applicable to the production of any protein of a heterologous nature. Among these proteins, mention may be made in particular of hirudin or defensins, for example defensin A.

The invention relates in particular to a process for the secretion of hirudin in mature form, starting from yeast strains which have been transformed according to the invention.

The invention is particularly applicable to the production of hirudin, which is why one of the illustrative examples of the invention concerns this protein. Hirudin, whose main source is found in the salivary glands of medicinal leeches, is in fact a very specific and very effective thrombin inhibitor. It is a very interesting therapeutic Agent whose clinical use requires a very high purity of the product and which is therefore an interesting candidate for production by genetic engineering.

A certain number of natural variants of hirudin have already been identified and are designated HV1, HV2, HV3. These natural variants and other analogues are produced below by genetic engineering in various host cells, as described for example in the applicant's European patent publications EP-A-0200655, EP-A-0273800. A comparison between hirudin synthesized by Escherichia coli (E. coli) and hirudin synthesized by a yeast of the genus S. cerevisiae has shown that hirudin synthesized by E. coli remains intracellular and thus has to be purified, starting from a very large number of E. coli polypeptides. It is therefore particularly advantageous (interesting) to be able to express a hirudin gene in yeast in such a way that a hirudin secreted in mature form is obtained without the yeast producing pyrogenic or toxic substances for humans.

The invention is applicable to all hirudin molecules, i.e. to all natural variants of hirudin as such or to those which have undergone one or more mutations while retaining their antithrombotic activity, the latter type of variant being referred to as analogues. The examples below relate in particular to the analogue referred to as rHV2Lys47 (for the recombinant variant HV2 which has undergone a mutation of the amino acid Asp at position 47 to the amino acid Lys), as described in the aforementioned European patent publication EP-A-0273800.

The present invention is also applicable to the production of defensins. Defensins, also called phormicins, are peptides originally derived from from the haemolymph of certain insects, the Diptera (Two-winged insects), which have a bactericidal activity against Gram-positive germs. These defensins are described in detail in the European patent application EP-A- 5349 451. Defensin A is a basic peptide that has the sequence Ala Thr Cys Asp Leu Leu Ser Gly Thr Gly Ile Asn His Ser Ala Cys Ala Ala His Cys Leu Leu Arg Gly Asn Arg Gly Gly Tyr Cys Asn Gly Lys Gly Val Cys Val Cys Arg Asn

Defensin B differs from defensin A only by the amino acid in position 32, in which an arginine replaces a glycine.

The following examples show further characteristics and advantages of the present invention. These examples are illustrated by the following figures:

- Figure 1 shows the schematic structure of the plasmid pTG2958.

- Fig. 2 shows the schematic structure of the vectors M13TG3839 and M13TG3841. For M13TG3839, the hatched rectangles at the ends correspond to M13TG103 and for M13TG3841 they correspond to M13TG3149.

- Figure 3 shows the schematic structure of the vector M13TG3845, where the hatched rectangle corresponds to the M13TG3149.

- Figure 4 shows the schematic structure of the plasmid pTG3828.

- Figure 5 shows the schematic structure of the plasmid pTG3864.

example 1 Construction of vectors for expression of hirudin: pTG3864, pTG3867, pTG3894 and pTG3884. A. Construction of the vector M13TG3845

Plasmid pTG2958 (Fig. 1) differs little from plasmid pTG1833 described in European patent publication EP-A-252854, which carries the coding sequence for rHV2Asp47. Plasmid pTG2958 does not contain the artificially introduced restriction site HindIII. The plasmid pTG2958 contains:

- a fragment of 1217 base pairs corresponding to the 5' region of the MFα1 gene (containing the promoter, the signal peptide coding sequence, the "pro" region and a sequence coding for the Lys-Arg peptide) and of 4 base pairs (BglIIº site, Klenow treatment),

- a fragment of 234 base pairs containing the complementary DNA of rHV2Lys47,

- a fragment of 243 base pairs comprising the yeast terminator PGK,

- the PvuII-EcoRI fragment of pBR322, which includes, among other things, the origin of replication of this plasmid and the gene for resistance to ampicillin (2292 base pairs),

- the EcoRI-HindIII fragment of yeast plasmid 2u (form B) containing the yeast gene LEU2 in the deleted form inserted into the PstI site,

- a fragment HindIII-Smal of the yeast gene URA3.

The NcoI-NcoI fragment of the vector pTG2958 carrying the sequences LEU2-d, 2 u and URA3 is replaced by the NcoI-NcoI fragment of pTG2800 as described in the European patent publication EP-A-0268501, which contains the sequences of the plasmid 2 u and the gene URA3 deleted from its promoter (URA3-d) to form pTG2877.

The vector M13TG3839 (Fig. 2) is derived from M13TG103 (M.P. Kieny et al. (1983) "Gene", 26, 91-99) in which the fragment HindIII-HindIII of pTG2877 has been introduced into the same site. A restriction site SalI is introduced into this vector downstream of the translation termination codon of the region encoding rHV2Lys47 by site-directed mutagenesis using the following oligonucleotide:

5' CAATGAAAAATGGTCGACTATCAATCATAG

to form M13TG3839 SalI. A restriction site SphI is then placed upstream of the expression cassette introduced by eliminating the sequence URA3-d by site-directed mutagenesis using the following oligonucleotide:

5' GACGGCCAGTGAATTGGCATGCTATTGATAAGATTTAAAG to form M13TG3840.

The vector M13TG131 (M.P. Kieny et al. (1983), "Gene", 26, 91-99) is cleaved by PstI, the ends are exposed by treatment with the Klenow fragment of polymerase DNA I, and then religated to itself to form M13TG3160. This vector is then cleaved by SmaI and EcoRV and then religated again to form M13TG3149.

The SphI-SalI fragment of M13TG3840 (as described above), carrying the cassette for expression of rHV2Lys47 (without transcription termination sequence), is inserted into the SphI-SalI site of M13TG3149 to form M13TG3841 (Fig. 2).

The sequence coding for the amino acids in positions 3 to 42 on the "pro" fragment of the MFα1 precursor is eliminated from M13TG3149 by site-directed mutagenesis and a SmaI restriction site is introduced using the following oligonucleotide:

5' GTCCGCATTAGCTGCTCCCCGGGTTATTGTTTATAAAT

to form what is referred to below as the deleted "pro" sequence. M13TG3842 is thus obtained. A BamHI restriction site, which destroys the ATG of the precursor of factor α, is introduced without this vector by directed mutagenesis with the following oligonucleotide:

5' AATATAAACGATTAAAAGGATCCGATTTCCTTCAATTTTTA

This gives M13TG3843. After phosphorylation, the following oligonucleotides

5' GATCCGTTTCTCTACTACAGTCGCTACTGCAGCTACTGCGCTATT GCAAAGTGATGATGTCAGCGATG 5'

and

5' TTTCACAGCCTCCAAGTTTCAGCTGCTCCC ACGTCGATGACGCGATAAAAAGTGTCGGAGGGTTCAAAGTCGACGAGGG 5'

inserted into the vector M13TG3843, cut with BamHI and SmaI to introduce the sequence XI without ATG. To restore ATG, the BamHI site is eliminated by site-directed mutagenesis using the following oligonucleotide:

5' AATATAAACGATTAAAAGAATGCGTTTCTCTACTACAGTC

forming the vector MI3TG3845 (Fig. 3), which comprises:

- the promoter of the gene MFα1, followed by a codon ATG as a fragment (a),

- the sequence XI as fragment (b)

- the sequence "pro" deleted from the gene MFα1, followed by codons coding for Lys-Arg, as fragment (b'),

- the sequence encoding rHV2Lys47 as a fragment (c), and

- part of the vector M13TG3149.

B. Construction of plasmid pTG3864

Plasmid pTG848 described in European patent publication EP-A-0252854 is digested with BglII, then religated to form pTG2886. The large HindIII-EcoRI fragment of pTG2886 is ligated in the presence of ligase T4 with the 2.1 kb HindIII-EcoRI fragment of plasmid pFL1 (S.A. Parent et al. (1985), "Yeast", 1, 83-138) carrying the sequence of plasmid 2 u of S. cerevisiae to form plasmid pTG2886 LEU2-d, URA3-d. The 0.9 kb HindIII fragment of the plasmid pTG2800, described in European patent publication EP-A-0258501 and carrying the URA3-d gene, is then inserted into the HindIII site of this plasmid to form pTG2886 URA3-d, delta LEU2-d. The SmaI-BglII fragment of M13TG131 (Kieny et al. (1983), "Gene" 26, 91-99), which has several restriction sites, is then introduced into this plasmid to form pTG3828 (Fig. 4), which comprises:

- the sequence of the URA3 gene deleted from its promoter (URA3-d),

- restriction sites originating from M13TG131, which allow the insertion of elements for the expression of a heterologous gene, in this case rHV2Lys47,

- the transcription terminator of the yeast gene PGK,

- a fragment of pBR322 that allows replication and selection in E. coli,

- a fragment of plasmid 2u containing the structural elements required for replication and mitotic division into equal parts in yeast.

The SphI-SalI fragment of the vector M13TG3845 (Fig. 3) is introduced into the plasmid pTG3828, digested with SphI and SalI to form the expression vector pTG3864 (Fig. 5).

C. Construction of the expression vector pTG3867

To completely suppress the sequence coding for the precursor of the MFα1 gene, a site-directed mutagenesis is performed on M13TG3845 (Fig. 3) using the following oligonucleotide:

5' GCCTCCCAAGTTTCAGCTATTACGTATACAGACTGC

obtaining the vector M13TG3846. The SphI-SalI fragment of this vector is introduced into the plasmid pTG3828 (Fig. 4) and digested with SphI and SalI to form the expression vector pTG3867. In this vector, the XI sequence and that coding for rHV2Lys47 are arranged adjacent to each other. To obtain the schematic structure of this plasmid, it is sufficient to remove the "pro" sequence deleted from MFα1 in Fig. 5.

D. Construction of the expression vector DTG3894

In the sequence encoding the "pro" fragment of the precursor of factor α, a site is created 5 times by site-directed mutagenesis on M13TG3841 (Fig. 2) using the following oligonucleotide:

5' TCCGCATTAGCTGCTCCCGGGAACACTACAACAGAA

to obtain M13TG3869. This vector is then digested with Sphl and SmaI, the small fragment is isolated and it is ligated with the large SphI and SmaI fragment of M13TG3845 (Fig. 3) to form the vector M13TG3891. The SphI-SalI fragment of this vector is introduced into the plasmid pTG3828 (Fig. 4) open at the SphI and SalI sites to form the expression vector pTG3894, which thus contains the mutated "pro" sequence of the precursor of factor α. To obtain the schematic structure of this plasmid, it is sufficient to replace the "pro" sequence deleted from MFα1 with the mutated "pro" sequence in Fig. 5.

E. Construction of the expression vector pTG3884

Sequence XI is modified by site-directed mutagenesis with M13TG3845 using the following oligonucleotide:

5' GTTTCTCTACTACACTCGCTACTGC

Modification of a single base gives the sequence XII and induces the replacement of a valine by a leucine as amino acid R4 of the signal peptide. The bacteriophage MI3TG3846 is thus obtained. The SphI-SalI fragment of M13TG3846 is introduced into the plasmid pTG3828 (Fig. 4), open at the SphI and SalI sites, to form the expression vector pTG3884, which comprises:

- the sequence of the URA3 gene deleted from its promoter (URA3-d),

- the promoter of the gene MFα1, followed by a codon ATG as a fragment (a),

- sequence XII as fragment (b),

- the sequence "pro" deleted from the gene MFα1, followed by codons coding for Lys-Arg, as fragment (b'),

- the sequence encoding rHV2Lys47 as a fragment (c),

- the terminator of the gene encoding yeast PGK,

- a fragment of pBR322,

- a fragment of plasmid 2 u.

Example 2 Formation of rHV2Lvs47 in the culture supernatant as a function of the plasmid used

A yeast strain of the species Saccharomyces cerevisiae of the genotype MATα, ura3-251,-373,-328, leu2-3,-112, his3, pep4-3 is transformed by the plasmids pTG3864, pTG3867, pTG3894 and pTG3884 according to the lithium acetate method (H. Ito et al., "J. Bacteriol." (1983), 153: 163) and the prototrophs Ura+ are selected. They are then cultured in an Erlenmeyer flask at 30ºC on a selective medium (0.7% yeast nitrogen base, 0.5% casamino acids and 1% glucose). After 48 hours of cultivation, the cells and the supernatant are separated by centrifugation and the thrombin inhibitory activity in the supernatant is determined using the colorimetric test (proteolytic activity on a synthetic substrate, Chromozyin TH - Boehringer Mannheim). The results of the analysis are given in Table I below; each value corresponds to the mean of two independent experiments. The activity of rHV2Lys47 is expressed in ATU/ml of supernatant. Table I

Plasmid

In all cases, antithrombin activity is measured. The protein rHV2Lys47 produced by the yeast is then excreted into the supernatant. In addition, it is secreted in active form. The best results are obtained with the strains transformed by pTG3884 and pTG3867.

The protein content of the supernatants is analyzed by HPLC. The main peak obtained corresponds to that of rhv2Lys47 (in its 65 amino acid form) and the determination of the N-terminal sequence confirms that a correctly synthesized molecule has been obtained.

Example 3 Construction of a vector for the expression of insect defensin A: pTG4826 A: Synthesis of a DNA sequence encoding insect defensin A

The synthesis is carried out in two blocks linked by their cohesive ends Kpnl. The first block comprises 3 oligonucleotides numbered 1 to 3 and the second block comprises 6 oligonucleotides numbered 4 to 9. Their sequence and the position of the oligonucleotides (last row of the table, the circles represent the 5-section of the oligonucleotide) are given in Table II. Table II No. Sequence

The resulting sequence is the following:

In the synthesis of the first block, oligonucleotides 4, 5, 6, 7, 8 and 9 are used and the synthesis is carried out as follows:

- oligonucleotides 5, 6, 7 and 8 are first phosphorylated at their 5' ends to avoid the formation of polymers during assembly. For each of these oligonucleotides, 100 pmol of polynucleotide kinase are treated, two units in a final volume of 20 µl of 60 µM Tris HCl at pH 7.5; 10 µM MgCl₂; 8 µM dithiothreitol (kination buffer) containing 3.3 pmol of ATPγ labeled with 32P (5000 Ci/mmol). After incubation for 15 minutes at 37°C, 5 µmol of unlabeled ATP are added;

- after incubation for 30 minutes at 37ºC, 75 pmol of oligonucleotides 5, 6, 7 and 8 are mixed together, heated to 95ºC for 3 minutes, then oligonucleotides 4 and 9 are added in a final volume of 90 pmol of the kination buffer described above. The whole is heated to 95ºC for 3 minutes, then slowly cooled to 37ºC over 2 hours;

25 pmol of these hybridized oligonucleotides are treated with ligase T4 for 1 h at 15°C. This reaction mixture (1 pmol) is then added to 50 ng of the bacteriophage MI3TGI3I (M.P. Kieny et al. (1983), "Gene", 26 91-99) and treated with EcoRI and KpnI (1 h at 15°C). The ligation mixture is used to transform competent cells of the E. coli JM103 strain (J. Messing et al. (1981), "Nucleic Acid Res.", 2, 309). A clone containing the sequence of interest is isolated and designated M13TG3821.

In the synthesis of the second block, oligonucleotides 1, 2 and 3 are used and the synthesis is carried out following the same procedure as described for the synthesis of the first block. In this case, only oligonucleotide 2 is phosphorylated at its 5'-end.

This second block is cloned between the HindIII and KpnI sites of the bacteriophage M13TG3821 and a clone carrying the DNA sequence encoding defensin A (Fig. 5) was isolated; it is designated MI3TG3849.

B. Construction of the plasmid for expression of defensin A: pTG4839.

The 1045 base pair SphI - SmaI fragment of the bacteriophage M13TG3846 described above (Example 1, E) is transferred into the vector MI3TG3869 described above (Example 1, D) which has been previously digested with SphI and SmaI. This gives the vector M13TG4803 which carries:

- the promoter of the gene MFal, followed by a codon ATG,

- the sequence XII,

- the mutated sequence "pro" of the MFal gene, followed by codons coding for Lys-Arg,

- the sequence encoding rHV2Lys47.

To replace the sequence coding for rHV2Lys47 with that coding for insect defensin A, a HindIII site is introduced into the mutated sequence of MFα1 coding for "pro" using the following oligonucleotide:

5' GAAGGGTAAGCTTGGATAAA

Then, the HindIII-BamHI fragment of M13TG3849 described above (Example 3, A), which carries the synthetic sequence coding for defensin A, is introduced into this vector, which has previously been treated with HindIII and BamHI to eliminate the sequence coding for rHV2Lys47.

The SphI-SalI fragment of M13TG4813 is introduced into the plasmid pTG3828 (Fig. 4) open at the SphI and SalI sites, to form the expression vector pTG4839, which comprises:

- the sequence of the URA3 gene deleted from its promoter (URA3-d),

- the promoter of the gene MFα1, followed by an ATG

- the sequence XII,

- the mutated sequence "pro" of the MFα1 gene, followed by codons coding for Lys-Arg,

- the synthetic sequence encoding insect defensin A,

- the terminator of the gene encoding yeast PGK,

- a fragment of pBR322,

- a fragment of plasmid 2u.

Example 4 Formation of defensin A in the culture supernatant

A yeast strain of the species Saccharomyces cerevisiae of the genotype MATα, ura3-251,-373,-328, leu2-3,-112, his3, pep4-3 is transformed by the plasmid pTG4839 using the lithium acetate method (H. Ito et al., "J. Bacteriol." (1983), 153: 163) and the prototrophs Ura+ are selected. They are then cultured in an Erlenmeyer flask at 30ºC on a selective medium (0.7% yeast nitrogen base, 0.5% casamino acids and 1% glucose). After 48 hours of cultivation, the cells and supernatant are separated by centrifugation and the supernatant is filtered through a 22 u filter, then passed through a Sep-Pak C18 cartridge. The fixed material is eluted with 60% acetonitrile, 0.1% trifluoroacetic acid in water and dried under vacuum. The antibacterial activity of defensin A is then demonstrated using a streak test on agar or on agar-agar inoculated with bacteria (Micrococcus luteus) according to the procedure described by Lambert et al. (1989) in "PNAS" 86: 262-266.

Antibacterial activity is effectively detected in the supernatants of yeasts transformed by the plasmid pTG4839. The protein defensin A produced by the yeast is thus excreted into the supernatant. In addition, it is secreted in the active form.

The protein content of the supernatants is analyzed by HPLC. The main peak obtained corresponds to that of insect defensin A and the determination of the protein sequence confirms that a correctly synthesized molecule has been obtained.

Claims (26)

1. Isolated DNA fragment encoding a signal peptide suitable for the secretion of a heterologous protein by a cell in which the heterologous protein is synthesized, the amino acid sequence of which has a degree of homology of at least 60% with the amino acid sequence of formula (I) or (II) 2. DNA fragment according to claim 1, which encodes a peptide whose amino acid sequence has a degree of homology of at least 80% with the amino acid sequence of formula (I) or (II) 3. DNA fragment according to claim 1 or 2, which encodes a peptide having the amino acid sequence (III) in the: R₁ represents an amino acid selected from Arg and Lys, R₂ and R₆ each represent an amino acid independently selected from Ala, Asn, Cys, Gln, Gly, His, Iie, Leu, Met, Phe, Pro, Ser, Thr, Trp Tyr and Val R₃ and R₅ each represent an amino acid independently selected from Asp, Gly, Asn, Pro and Ser and R4 represents an amino acid selected from Val, Leu, Ala, Cys, Phe, Ile and Met. 4. DNA fragment according to claim 1 or 2, which encodes a peptide having the amino acid sequence (IV) in the: R₁ represents an amino acid selected from Arg and Lys, R₂ and R₆ each represent an amino acid independently selected from Ala, Asn, Cys, Gln, Gly, His, Ile, Leu, Met, Phe, Pro, Ser, Thr, Trp, Tyr and Val R₃ and R₅ each represent an amino acid independently selected from Asp, Gly, Asn, Pro and Ser, R₄ represents an amino acid selected from Val, Leu, Ala, Cys. Phe, Leu and Met, and R�7 stands for a proteolysis sequence. 5. DNA fragment according to claim 4, in which R₇ is a proteolysis sequence R₈-R₆-R₁₀ in which: R�8 represents an amino acid selected from Ala, Val, Ser, Cys, Gly, Ile, Leu and Thr, R�9 represents an amino acid selected from Ala, Arg, Cys, Gln, Gly, His, Ile, Leu, Met, Phe, Pro, Ser, Ihr, Trp, Tyr and Val and R₁₀ represents an amino acid selected from Ala, Cys, Gly, Leu, Pro, Gln, Ser and Thr. 6. DNA fragment according to claim 1 or 2, which encodes a peptide containing an amino acid sequence selected from the amino acid sequences (V) and (VI) 7. DNA fragment according to any one of claims 1, 2, 4 and 5, which encodes a peptide having an amino acid sequence selected from the amino acid sequences of formula (VII) and (VIII) in which R7 represents a proteolysis sequence; in which R7 represents a proteolysis sequence. 8. DNA fragment according to claim 7, which encodes a peptide having an amino acid sequence selected from the amino acid sequences of formula (VII) and (VIII) in which R7 is Val-Ser-Ala. 9. Use of a DNA fragment according to one of claims 1 to 8, characterized in that the DNA fragment codes for a signal peptide which is suitable for the secretion of a heterologous protein by a yeast cell in which the heterologous protein is synthesized. 10. Cassette for the expression of a heterologous protein, comprising: a) a DNA fragment containing transcription and translation initiation signals, b) a DNA fragment according to any one of claims 1 to 8 and c) a DNA fragment encoding the mature heterologous protein. 11. Cassette according to claim 10, further comprising a DNA fragment (b') encoding a peptide fragment "pro". 12. Cassette according to claim 11, which comprises a DNA fragment (b') which encodes a peptide fragment "pro" having the sequence Ma Pro Gly Leu Leu Phe Ile Asn Thr Thr Ile Ala Ser Ile Ma Ma Lys Glu Glu Gly Val Ser Leu Asp Lys Arg. 13. Cassette according to one of claims 10 to 12, characterized in that the fragment (a) has a functional promoter in a yeast cell and a translation initiation codon ATG. 14. Expression cassette according to one of claims 10 to 13, characterized in that the DNA fragment (c) codes a hirudin. 15. Expression cassette according to claim 14, characterized in that the DNA fragment (c) codes for the hirudin variant rHV2Lys47. 16. Expression cassette according to one of claims 10 to 13, characterized in that the DNA fragment (c) codes an insect defensin. 17. Expression cassette according to claim 16, characterized in that the DNA fragment (c) encodes defensin A. 18. Plasmid vector containing an expression cassette according to any one of claims 10 to 17. 19. Plasmid vector according to claim 18, characterized in that it contains a fragment of the yeast plasmid 2 u. 20. Plasmid vector according to claim 18 or 19, characterized in that it contains as selection gene the gene URA3 from which its promoter has been deleted. 21. Cell which has been transformed by a plasmid vector according to one of claims 18 to 20 or into whose genome an expression cassette according to one of claims 10 to 17 has been integrated. 22. Yeast cell according to claim 21. 23. A process for producing a heterologous protein, characterized in that a cell according to claim 21 or 22 is cultivated and that said protein is isolated from the culture medium. 24. Method according to claim 23, characterized in that the cell is a yeast cell. 25. Method according to claim 23 or 24, characterized in that the said protein is a hirudin. 26. A method according to claim 23 or 24, characterized in that said protein is an insect defensin.